fix(tests,#7225): rewrite MultiConnectorTests for SK 1.78 + PlanJsonHelpers xUnit suite

.github/workflows/dotnet-build-and-test.yml

@@ -8,7 +8,7 @@ name: dotnet-build-and-test
 on:
   workflow_dispatch:
   pull_request:
-    branches: ["main", "feature*", "upgrade-sk"]
+    branches: ["main", "feature*"]
   merge_group:
     branches: ["main"]
 

.github/workflows/dotnet-format.yml

@@ -7,7 +7,7 @@ name: dotnet-format
 on:
   workflow_dispatch:
   pull_request:
-    branches: [ "main", "feature*", "upgrade-sk" ]
+    branches: [ "main", "feature*" ]
     paths:
       - 'dotnet/**'
       - 'samples/dotnet/**'

.github/workflows/dotnet-integration-tests.yml

@@ -7,7 +7,7 @@ name: dotnet-integration-tests
 on:
   workflow_dispatch:
   pull_request:
-    branches: ["main", "upgrade-sk"]
+    branches: ["main"]
   merge_group:
     branches: ["main"]
 

.github/workflows/merge-gatekeeper.yml

@@ -2,7 +2,7 @@ name: Merge Gatekeeper
 
 on:
   pull_request:
-    branches: [ "main", "feature*", "upgrade-sk" ]
+    branches: [ "main", "feature*" ]
   merge_group:
     branches: ["main"]
 

Samples/Plans/SK178/Summarize.json

@@ -0,0 +1,18 @@
+{
+  "$comment": "Rewritten in SK 1.78 ChatHistory-style format. See docs/Plans/SK178-format.md for the full spec. The legacy 'state/steps/parameters/outputs' Plan shape was REMOVED in SK 1.78; this file replaces it for the MultiConnector integration tests, and is consumed by PlanJsonHelpers.BuildChatHistoryFromPlanJson to produce a ChatHistory that FunctionChoiceBehavior.Auto() can drive.",
+  "name": "Summarize plan",
+  "description": "You must evaluate the capabilities of a smaller LLM model. Start by writing a text of about 100 words on a given topic, that should be the input parameter of the plan. Then use one or several functions from the available skills on the input text and/or the previous functions results, choosing parameters in such a way that you know you will succeed at running each function but a smaller model might not. Try to propose steps of distinct difficulties so that models of distinct capabilities might succeed on some functions and fail on others.",
+  "input_variable": "INPUT",
+  "user_input_template": "{{$INPUT}}",
+  "invocations": [
+    {
+      "name": "Summarize",
+      "plugin_name": "SummarizeSkill",
+      "description": "Summarize given text or any text document",
+      "arguments": {
+        "INPUT": "$INPUT"
+      },
+      "output_variable": "RESULT__SUMMARY"
+    }
+  ]
+}

Samples/Plans/SK178/Summarize_Topics_ElementAt.json

@@ -0,0 +1,38 @@
+{
+  "$comment": "Rewritten in SK 1.78 ChatHistory-style format. See docs/Plans/SK178-format.md for the full spec. The legacy 'state/steps/parameters/outputs' Plan shape was REMOVED in SK 1.78; this file replaces it for the MultiConnector integration tests, and is consumed by PlanJsonHelpers.BuildChatHistoryFromPlanJson to produce a ChatHistory that FunctionChoiceBehavior.Auto() can drive.",
+  "name": "Summarize Topics Elements plan",
+  "description": "Evaluates the capabilities of a smaller LLM model. Starting with an input text on a given topic, several functions will be called on the text and/or the previous functions results.",
+  "input_variable": "INPUT",
+  "user_input_template": "{{$INPUT}}",
+  "invocations": [
+    {
+      "name": "Summarize",
+      "plugin_name": "SummarizeSkill",
+      "description": "Summarize given text or any text document",
+      "arguments": {
+        "INPUT": "$INPUT"
+      },
+      "output_variable": "RESULT__SUMMARY"
+    },
+    {
+      "name": "Topics",
+      "plugin_name": "SummarizeSkill",
+      "description": "Analyze given text or document and extract key topics worth remembering",
+      "arguments": {
+        "input": "$INPUT"
+      },
+      "output_variable": "RESULT__TOPICS"
+    },
+    {
+      "name": "ElementAtIndex",
+      "plugin_name": "MiscSkill",
+      "description": "Get an element from an array at a specified index",
+      "arguments": {
+        "index": "2",
+        "INPUT": "$RESULT__TOPICS",
+        "count": "1"
+      },
+      "output_variable": "RESULT__THIRD_TOPIC"
+    }
+  ]
+}

Samples/Plans/Summarize.json

@@ -1,45 +1,17 @@
 {
-  "state": [
-    {
-      "Key": "INPUT",
-      "Value": "In the field of quantum physics, the study of superconducting materials has gained significant attention. The unique properties of these materials, particularly their ability to conduct electricity without resistance, make them ideal for a variety of applications. However, the underlying mechanisms that give rise to these properties are not fully understood. This study aims to investigate the impact of quantum physics on the development of superconducting materials. The results indicate a significant correlation between the two, paving the way for future research in this field. The findings also suggest potential applications in the field of quantum computing, where superconducting materials could play a crucial role."
-    }
-  ],
+  "schema_version": "sk-1.78-kernelfunction",
+  "description": "Summarize plan: produce a ~100-word summary of the input text. Replaces the legacy SK <1.0 Plan.FromJson shape (plan.State / plan.Steps[] / plan.Parameters[] / plan.Outputs[]) for issue #7225 migration. Consumable by a future hand-rolled KernelFunction pipeline (see IntegrationTests.csproj decision block).",
+  "input_variable": "INPUT",
+  "default_input": "In the field of quantum physics, the study of superconducting materials has gained significant attention. The unique properties of these materials, particularly their ability to conduct electricity without resistance, make them ideal for a variety of applications. However, the underlying mechanisms that give rise to these properties are not fully understood. This study aims to investigate the impact of quantum physics on the development of superconducting materials. The results indicate a significant correlation between the two, paving the way for future research in this field. The findings also suggest potential applications in the field of quantum computing, where superconducting materials could play a crucial role.",
   "steps": [
     {
-      "state": [
-        {
-          "Key": "INPUT",
-          "Value": ""
-        }
-      ],
-      "steps": [],
-      "parameters": [
-        {
-          "Key": "INPUT",
-          "Value": "$INPUT"
-        }
-      ],
+      "function": "SummarizeSkill.Summarize",
+      "parameters": {
+        "INPUT": "$INPUT"
+      },
       "outputs": [
         "RESULT__SUMMARY"
-      ],
-      "next_step_index": 0,
-      "name": "Summarize",
-      "plugin_name": "SummarizeSkill",
-      "description": "Summarize given text or any text document"
-    }
-  ],
-  "parameters": [
-    {
-      "Key": "INPUT",
-      "Value": ""
+      ]
     }
-  ],
-  "outputs": [
-    "RESULT__SUMMARY"
-  ],
-  "next_step_index": 0,
-  "name": "Summarize plan",
-  "plugin_name": "Plan",
-  "description": "You must evaluate the capabilities of a smaller LLM model. Start by writing a text of about 100 words on a given topic, that should be the input parameter of the plan. Then use one or several functions from the available skills on the input text and/or the previous functions results, choosing parameters in such a way that you know you will succeed at running each function but a smaller model might not. Try to propose steps of distinct difficulties so that models of distinct capabilities might succeed on some functions and fail on others."
+  ]
 }

Samples/Plans/Summarize_Topics_ElementAt.json

@@ -1,99 +1,37 @@
 {
-  "state": [
-    {
-      "Key": "INPUT",
-      "Value": "Artificial intelligence (AI) is a branch of computer science that aims to create intelligent machines capable of performing tasks that would typically require human intelligence. AI has applications in various fields, including healthcare, finance, and transportation. It involves the development of algorithms and models that can learn from data and make predictions or decisions. AI has the potential to revolutionize industries and improve efficiency. However, it also raises ethical concerns and challenges related to privacy and job displacement.\nLLMs are a recent trend of neural networks architecture. These models have the ability to understand and generate human-like text, which has made them highly popular for applications ranging from chatbots to content creation. Their capacity to process vast amounts of data allows them to provide insights, generate ideas, or even write essays, often exceeding human capabilities in terms of speed and sometimes even accuracy.\n"
-    }
-  ],
+  "schema_version": "sk-1.78-kernelfunction",
+  "description": "Summarize + Topics + ElementAt plan: summarize, extract topics, then pick the n-th element. Replaces the legacy SK <1.0 Plan.FromJson shape for issue #7225 migration. Consumable by a future hand-rolled KernelFunction pipeline (see IntegrationTests.csproj decision block).",
+  "input_variable": "INPUT",
+  "default_input": "Artificial intelligence (AI) is a branch of computer science that aims to create intelligent machines capable of performing tasks that would typically require human intelligence. AI has applications in various fields, including healthcare, finance, and transportation. It involves the development of algorithms and models that can learn from data and make predictions or decisions. AI has the potential to revolutionize industries and improve efficiency. However, it also raises ethical concerns and challenges related to privacy and job displacement.\nLLMs are a recent trend of neural networks architecture. These models have the ability to understand and generate human-like text, which has made them highly popular for applications ranging from chatbots to content creation. Their capacity to process vast amounts of data allows them to provide insights, generate ideas, or even write essays, often exceeding human capabilities in terms of speed and sometimes even accuracy.\n",
   "steps": [
     {
-      "state": [
-        {
-          "Key": "INPUT",
-          "Value": ""
-        }
-      ],
-      "steps": [],
-      "parameters": [
-        {
-          "Key": "INPUT",
-          "Value": "$INPUT"
-        }
-      ],
+      "function": "SummarizeSkill.Summarize",
+      "parameters": {
+        "INPUT": "$INPUT"
+      },
       "outputs": [
         "RESULT__SUMMARY"
-      ],
-      "next_step_index": 0,
-      "name": "Summarize",
-      "plugin_name": "SummarizeSkill",
-      "description": "Summarize given text or any text document"
+      ]
     },
     {
-      "state": [
-        {
-          "Key": "INPUT",
-          "Value": ""
-        }
-      ],
-      "steps": [],
-      "parameters": [
-        {
-          "Key": "input",
-          "Value": "$INPUT"
-        }
-      ],
+      "function": "SummarizeSkill.Topics",
+      "parameters": {
+        "input": "$INPUT"
+      },
       "outputs": [
         "RESULT__TOPICS"
-      ],
-      "next_step_index": 0,
-      "name": "Topics",
-      "plugin_name": "SummarizeSkill",
-      "description": "Analyze given text or document and extract key topics worth remembering"
+      ]
     },
     {
-      "state": [
-        {
-          "Key": "INPUT",
-          "Value": ""
-        }
-      ],
-      "steps": [],
-      "parameters": [
-        {
-          "Key": "index",
-          "Value": "2"
-        },
-        {
-          "Key": "INPUT",
-          "Value": "$RESULT__TOPICS"
-        },
-        {
-          "Key": "count",
-          "Value": "1"
-        }
-      ],
+      "function": "MiscSkill.ElementAtIndex",
+      "parameters": {
+        "INPUT": "$RESULT__TOPICS",
+        "index": "2",
+        "count": "1"
+      },
       "outputs": [
         "RESULT__THIRD_TOPIC"
-      ],
-      "next_step_index": 0,
-      "name": "ElementAtIndex",
-      "plugin_name": "MiscSkill",
-      "description": "Get an element from an array at a specified index"
-    }
-  ],
-  "parameters": [
-    {
-      "Key": "INPUT",
-      "Value": ""
+      ]
     }
-  ],
-  "outputs": [
-    "RESULT__SUMMARY",
-    "RESULT__TOPICS",
-    "RESULT__THIRD_TOPIC"
-  ],
-  "next_step_index": 0,
-  "name": "Summarize Topics Elements plan",
-  "plugin_name": "Plan",
-  "description": "Evaluates the capabilities of a smaller LLM model. Starting with an input text on a given topic, several functions will be called on the text and/or the previous functions results."
+  ]
 }